RE: [PATCH net-next 00/13] resource management using devlink reload

From: "Saleem, Shiraz" <shiraz.saleem@intel.com>
To: Leon Romanovsky <leon@kernel.org>, Jakub Kicinski <kuba@kernel.org>
Cc: Michal Swiatkowski <michal.swiatkowski@linux.intel.com>,
	"Samudrala, Sridhar" <sridhar.samudrala@intel.com>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	"davem@davemloft.net" <davem@davemloft.net>,
	"pabeni@redhat.com" <pabeni@redhat.com>,
	"edumazet@google.com" <edumazet@google.com>,
	"intel-wired-lan@lists.osuosl.org"
	<intel-wired-lan@lists.osuosl.org>,
	"jiri@nvidia.com" <jiri@nvidia.com>,
	"Nguyen, Anthony L" <anthony.l.nguyen@intel.com>,
	"Lobakin, Alexandr" <alexandr.lobakin@intel.com>,
	"Drewek, Wojciech" <wojciech.drewek@intel.com>,
	"Czapnik, Lukasz" <lukasz.czapnik@intel.com>,
	"Brandeburg, Jesse" <jesse.brandeburg@intel.com>,
	"Ismail, Mustafa" <mustafa.ismail@intel.com>,
	"Kitszel, Przemyslaw" <przemyslaw.kitszel@intel.com>,
	"Raczynski, Piotr" <piotr.raczynski@intel.com>,
	"Keller, Jacob E" <jacob.e.keller@intel.com>,
	"Ertman, David M" <david.m.ertman@intel.com>,
	"Kaliszczuk, Leszek" <leszek.kaliszczuk@intel.com>,
	Bjorn Helgaas <bhelgaas@google.com>
Subject: RE: [PATCH net-next 00/13] resource management using devlink reload
Date: Fri, 18 Nov 2022 14:23:50 +0000	[thread overview]
Message-ID: <MWHPR11MB002998FAA385731E21E20868E9099@MWHPR11MB0029.namprd11.prod.outlook.com> (raw)
In-Reply-To: <Y3ckRWtAtZU1BdXm@unreal>

> Subject: Re: [PATCH net-next 00/13] resource management using devlink reload
> 
> On Thu, Nov 17, 2022 at 07:36:18PM -0800, Jakub Kicinski wrote:
> > On Thu, 17 Nov 2022 19:38:48 +0200 Leon Romanovsky wrote:
> > > I don't think that management of PCI specific parameters in devlink
> > > is right idea. PCI has his own subsystem with rules and assumptions,
> > > netdev shouldn't mangle them.
> >
> > Not netdev, devlink, which covers netdev, RDMA and others.
> 
> devlink is located in net/*, it is netdev.
> Regarding RDMA, it is not fully accurate. We use devlink to convey information to
> FW through pci device located in netdev. Some of such parameters are RDMA
> related. However, we don't configure RDMA properties through devlink, there is a
> special tool for that (rdmatool).

rdmatool though is usable only once the rdma driver probe() completes and the ib_device is registered.
And cannot be used for any configurations at driver init time.

Don't we already have PCI specific parameters managed through devlink today?

https://docs.kernel.org/networking/devlink/devlink-params.html
enable_sriov
ignore_ari
msix_vec_per_pf_max
msix_vec_per_pf_min

How are these in a different bracket from what Michal is trying to do? Or were these ones a bad idea in hindsight?

Shiraz